Friday, 2017-02-10

*** gcb has quit IRC00:10
*** dave-mccowan has joined #openstack-python301:11
*** gcb has joined #openstack-python301:56
*** dave-mccowan has quit IRC03:42
*** tuanluong has joined #openstack-python303:56
*** tuanluong has quit IRC04:35
*** gcb has quit IRC04:51
*** gcb has joined #openstack-python304:53
*** gcb has quit IRC05:29
*** gcb has joined #openstack-python305:30
*** gcb has quit IRC06:18
*** gcb has joined #openstack-python306:18
*** tuanluong has joined #openstack-python307:45
*** tuanluong has quit IRC10:44
gcbhaypo, only these two: https://review.openstack.org/#/q/status:open+project:openstack/nova+branch:master+topic:bug/148263312:06
haypogcb: did you get support from nova cores?12:07
*** dave-mccowan has joined #openstack-python312:08
haypogcb: oh... SSL. not sure if something is wrong in eventlet, or if it's because ssl became very strict in python312:08
gcbhaypo, yeah, can't do that with their support :-)12:08
haypogcb: but ssl also became stricter in python2, but it depends in the minor 2.7.x version12:08
gcbhaypo, we did same thing to skip unittest in oslo.service, nova just follow the way12:08
gcbhttps://github.com/openstack/oslo.service/blob/master/oslo_service/tests/test_wsgi.py#L28212:10
gcbs/with/without12:11
haypogcb: yeah i know. it would be "nice" to dig into the issue, but i tried once or twice and failed to see something obvious :-/12:12
gcbhaypo: I also tried :-/12:15
haypogcb: yeah, it's painful to debug SSL12:29
haypogcb: that's why it's also to wait until aaaaall other issues are fixed before looing at this one :-D12:30
gcbhaypo: agree :-)12:31
*** dave-mccowan has quit IRC12:32
*** dave-mccowan has joined #openstack-python312:33
*** gcb has quit IRC12:55
*** frickler has joined #openstack-python313:07
fricklerif anyone has time to look at https://bugs.launchpad.net/tempest/+bug/1663593 that would be nice.13:13
openstackLaunchpad bug 1663593 in tempest "py35 dsvm tests failing in nova novnc" [Undecided,New]13:13
haypofrickler: you should probably replace '<html>' with b'<html>'13:23
fricklerhaypo: that would then end up like this: http://paste.openstack.org/show/598404/ do you think this looks reasonable or do you think there is something more elegant?13:45
haypofrickler: i don't think that b'Host: %s:%s\r\n' % (url.hostname, url.port) works on Python 3 using python3 -bb13:46
fricklerhaypo: o.k., so looking at https://wiki.openstack.org/wiki/Python3#bytes.decode_and_unicode.encode I would then use reqdata.encode('utf-8') in the end instead?14:02
haypofrickler: you have two choices for '<html>' in resp.data14:03
haypofrickler: (1) b'<html>' in resp.data14:03
haypofrickler: (2) '<html>' in resp.data.decode('utf-8')14:03
haypofrickler: (2) works in more cases, according to your change14:03
fricklerhaypo: so I'm trying with this one, will see what the CI has to say about it https://review.openstack.org/43232114:09
fricklerI was too late, https://review.openstack.org/#/c/429949/5 is already a couple of steps ahead :)14:25
*** dims_ has quit IRC15:28
*** dims has joined #openstack-python315:33
*** dims has quit IRC15:55
*** dims has joined #openstack-python315:55
*** dims is now known as Guest1240615:56
*** Guest12406 has quit IRC16:05
*** dims_ has joined #openstack-python316:24
*** dave-mccowan has quit IRC17:05
*** cdent has joined #openstack-python317:45
*** cdent has quit IRC18:34
*** dave-mccowan has joined #openstack-python320:12
*** dave-mccowan has quit IRC22:12
*** haypo has left #openstack-python322:13

Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!